Journal of the Royal Society of Medicine Volume 84 August 1991 451
Editorials 29 JUL 1991 Does malaria select for predisposition to autoimmune disease?
Over 20 years ago Dr Brian Greenwood observed that the admissions lists of a hospital in Nigeria contained very few cases of patients with autoimmune disease'. Of 98 454 admissions, only 104 were classified as autoimmune disease, including two with systemic lupus erythematosus (SLE) and 42 with rheumatoid arthritis, considerably less thanexpected (4- and 6-fold respectively) on the basis of figures for European hospitals. Greenwood suggested that parasitic disease, especially malaria, may in some way prevent the onset of autoimmune conditions. In contrast to its relatively low incidence in West Africans, SLE is more common in blacks (mainly descended from West Africans) than in whites in the USA2. The initial findings in West Africa were followed up by experimental work in which rodent malaria was shown to prevent the 'lupus-like' syndrome in one of the inbred strains of mice that spontaneously develop this symptom34 These interesting and straightforward observations have been largely ignored, probably because of the lack of a conceptual framework that could embrace the immune mechanisms involved in such complex events as malaria and autoimmunity. Recent advances in our understanding of the pathology of malaria and of factors that predispose to autoimmue disease (particularly SLE), suggest that a possible link between these conditions is tumour necrosis factor (TNF-aP. Although environmental factors are thought to precipitate SLE, there is a strong genetic prediosition in people with MHC class II DR2, DQwl and DR3 antigen alleles6. Mononuclear cells from subjects with DR2 and DQwl have a poor ability to produce TNF in vitro, suggesting-that this qytokine is involved in the pathogenesis of the disease6. However, SLE is also associated with DR3 aid mononuclear cells from donors with this specificity are good TNF producers6, the relationship is not a simple one. This is not surprising as SLE patients exhibit a broad spectrum of symptoms and the disease is considered to be an aggregate of similar clinical patterns rather than a single disease entity7. Nevertheless, it is significant that DR3 subjects do not exhibit the nephritis characteristic of other forms of SLE, while the DR2 and DQwl subjects do have this condition6. Further support for the argument that TNF- is involved in SLE-induced nephritis comes from experimental models: mice which are poor TNF producers and which are predisposed to lupus nephritis are by injections of TNF*. (Interestingly, a similar result is obtained with autoimmune diabetes9.) Malaria is a potent inducer of TNF in ai ,nals and man and it is now recognized that much of the pathology results from the initial induction of this cytokine'0. Patients infected with Plasmodium falciparum who develop very high plasma levels of so
TNF may be at greater risk from the potentially fatal condition of cerebral malariall. Malaria parasites express antigens (the so-called malaria 'toxins') which trigger macrophage release of TNF by T-cellindependent mechanisms'2, in a manner similar to endotoxin. Although the TNF-induced activation of macrophages and neutrophils enables the host to control parasite replication in the early stages,of the infection'3, there seems to be an uncontrolled overproduction of the cytokine that leads to a variety of pathological sequelae. This profound induction of TNF by malaria presumably explains why rodent malaria prevented the onset of autoimmune nephritis in susceptible strains of mice3'4 (and, possibly, adjuvant--nduced arthritis in rats'4). However, it would clearly be advantageous to the human host not to develop very high levels of TNF, thereby reducing the pathological consequences of malaria, and most especially the risk of cerebral symptoms. One might, therefore, expect to find that the long experience of malaria of populations in endemic areas, such as sub-Saharan Africa, would result in them exhibiting a lower level of symptoms than Caucasians. There is some evidence that this is indeed the case'5, although it has to be kept in mind that people from these areas may carry a variety, of inherited potective factors. It is conceivable that a consequence of a reduced ability to generate TNF in a malaria-free environment gives8rise to a greater risk of developing autoimmune conditions such as some forms of SILE - hence the higher rate of this in blacks in the USA (see above). It is well known that malaria may have acted as a selective agent for a variety of haemoglobinopathies etc., so it would not be suring if it has also selected genotypes involved in acute/non-specific immune responses. Unfortunately, our knowledge of the MHC genotypes of people in malaria endemic areas is rather limited, although the DQwl and DR2 haplotypes (see above) are slightly more commnn in blacks than in whites in the -USA (Hurley, personal-
communication). SLE, as already indicated, is a complex disease, considerably more common in women than men and usually with its onset during the main reproductive period. There may -be a further connection with malaria in this context, (although any attempt to explain the mechani involved at this stage has to be somewhat speculative). Even women who have become immune to malaria lose some (though not all) of this acquired resistnce during pregnancy and develop significant parasite burdens. The placenta becomes heavily parasitized, with inevitable consequences for the fetus in terms of reduced birth weight or possible abortion6L7. -However, the sytem by. stimulation of the mothefrs immu ye malaria may at the same time generate an increased output -of TNF and lessen the chances of SLE developing. Under natural conditions immunity to malaria is never complete, immune subjects always carry
0141-07/91/ 08041-W3=).00/0 © i99 The Royal Society of Medicine
452
Journal of the Royal Society of Medicine Volume 84 August 1991
parasites and suffer some fevers to a greater or lesser degree throughout their lives'8 so periodic minor boosts to the immune system are bound to occur and may be accompanied by rises in serum TNF19. It is especially pertinent to the argument that TNF administered to mice over a long period is immunosuppressive with respect to cell-mediated immunity (T cell proliferation, delayed hypersensitivity, etc.) but leaves humoral immunity largely unaffected20. The strain and species specificity of immunity to malaria, and the absence of delayed hypersensitivity to blood stage antigens21, suggests that naturally acquired resistance in humans is predominantly antibodymediated once it is established22, so the malaria itself is largely controlled and mortality is prevented. However, cell-mediated immunity, as exemplified by non-specific lymphocyte responses, is sometimes observed to be suppressed in subjects in endemic areas23. This mild general immunosuppression could perhaps contribute to preventing the loss of immune control characteristic of autoimmunity, by analogy with immunosuppressive drugs. Although malaria-induced TNF presumably contributes to the higher abortion rate of some mothers'7, (TNF induces abortion but a direct connection between malaria-induced abortion and TNF has yet to be confirmed) those generating low levels of TNF would have a greater chance of producing offspring, thus perpetuating their genotypes. Selection would be especially important at this point. Children in malaria-endemic countries carrying heavy burdens ofparasites without suffering apparent illness may develop antibodies to the parasite 'toxins' that directly stimulate TNF generation24, but it is also possible that their macrophages become tolerant to stimulation by these parasite components by analogy with endotoxin tolerance'3. Malaria could thus additionally select for the rapid development of control mechanisms that would shut down macrophage activation25 and TNF production, possibly through generation of other cytokines, particularly IL-426 and IL-627, as well as stimulating antimalarial antibody production28. In this context it may be significant that the genes for TNF map within the MHC complex6. Anecdotal evidence for malaria acting in an 'antiinflammatory' manner can be found in older literature. For example, P. vivax has been used to treat the gangrenous lesions of Buerger's disease29, thought by some physicians to result from an allergic sensitization to cigarette smoke; mention is also made of its use in the treatment of; other autoimmune problems29. Whatever the reliability of the data on Buerger's disease, a delicate balance may exist between the inflammatory and healing properties of cytokines, as proposed for IL-6 in the context of burns30, another condition in which acute nonspecific responses are generated. Although an understanding of autoimmune T cell reactivity3' is of major importance in explaining autoimmune disease (in addition to other factors, such as complement) a wider view that includes a focus onthe macrophage and TNF production, and important human parasites like malaria, may provide some useful clues that until now-have been ignored. Lastly, there are various grounds for thinking that a future malaria vaccine that allows the parasite to persist may be of more long-term benefit than one
which totally eliminates it from endemic areas. Members of these populations with a predisposition to autoimmune disease may be less disadvantaged by such a non-sterilizing vaccine. G A Butcher Department of Biology Imperial College of Science Technology and Medicine, Prince Consort Road, London SW7 2BB References 1 Greenwood BM. Autoimmune disease and parasitic infections in Nigerians. Lancet 1968;i:380-2 2 Lee SL, Siegel M. Systemic lupus erythematosus: epidemiological clues to pathogenesis. In: Dumonde D, ed. Infection and immunology in the rheumatic diseases. Oxford: Blackwell, 1976:307-17 3 Greenwood BM, Voller A. Suppression of autoimmune disease in New Zealand mice associated with infection with malaria. Clin Exp Immunol 1970;7:793-803 4 Greenwood BM, Herrick EM, Voller A. Suppression of autoimmune disease in NZB and (NZBxNZW) F1 hybrid mice by infection with malaria. Nature 1970;226:266-7 5 Butcher GA, Clark IA. SLE and malaria: another look at an old idea. Parasitol Today 1990;6:259-61 6 Jacob CO, Fronek ZL, Weis GD, et al. Proc Natl Acad Sci USA 1990;87:1233-7 7 Smolen JS, Chused TM, Leiserson WM et al. Heterogeneity of immunoregulatory T-cell subsets in systemic lupus erythematosus. Am J Med 1982;72:783-90 8 Jacob CO, McDevitt HO. Tumour necrosis factor-a in murine autoimmune 'lupus' nephritis. Nature 1988; 331:356-8 9 Satoh JO, Seino H, Shigeki S, et aL Inhibition of type 1 diabetes in BB rats with recombinant human tumor necrosis factor-a. J Immunol 1990;145:1395-9 10 Clark A, Chaudri G, Cowden WB. Role of tumour necrosis factor in the illness and pathology of malaria. Trans R Soc Trop Med Hyg 1989;83:436-46 11 Kwiatowski D, Hill AVS, Sambou I, et al. TNF concentration in fatal cerebral, non-fatal cerebral, and uncomplicated Plasmrnodium falciparum malaria. Lancet 1990;336:1201-4 12 Taverne J, Bate CAW, Sarkar DA, et al. Human and murine macrophages produce TNF in response to soluble antigens of Plasmodium falciparum. Parasie Immunol 1990;12:33-43 13 Clark IA. Cell-mediated immunity in protection and pathology of malaria. Parasitol Today 1987;3:300-5 14 Greenwood BM, Coller A, Herrick EM. Suppression of adjuvant arthritis by infection with a strain of the rodent malaria parasite Plasmodium berghei Ann Rheum Dis
1970;29:321-3 15 Powell RD, McNamara JV, Reickmann, KH. Clinical aspects of acquisition of immunity to falciparum malaria. Proc Helminth Soc Washington 1972;39(special
issue):51-66 16 Watkinson M, Rushton DI. Plasmodial pigmentation of placenta and outcome of pregnancy in West African mothers. BMJ 1983;287:251-4 17 McGregor IA. Epidemiology, malaria and pregnancy. Am J Trop Med Hyg 1984;33:517-25 18 Bruce-Chwatt LJ. A longitudenal survey of natural malaria infection in a group of West African adults. W Afr Med J 1963;12:141-73 19 Peyron F, Vuillez JP, Barbe G, et al Plasma levels of tumor necrosis factor during a longitudenal survey in an endemic area of malaria. Acta Trop 1990;47:47-51 20 Gordon C, Wofsy D. Effects of recombinant murine -tumor necrosis factor-a on immune function. JImmunol
1990*,l44:1753-8 21 Butcher~GA. In vitro responses of human peripheral blood mononuclear cells to Plasmodium fakciparum antigen. Int J Parasitol 1990;210:211-16
Journal of the Royal Society of Medicine Volume 84 August 1991 22 Butcher GA. Mechanisms of immunity to malaria and the possibilities of a blood stage vaccine: a critical appraisal. Parasitology 1989;98:315-27 23 Goonewardene R, Carter R, Gamage CP, et aL Human T cell proliferative responses to Plasmodium vivax: evidence of immunosuppression following prolonged exposure to endemic malaria. Eur J Immunol 1990;20:1201-16 24 Taverne J, Bate CAW, Playfair JHL. Malaria exoantigens induce TNF, are toxic and are blocked by Tindependent antibody. Immunol Lett 1990:25:207-12 25 Brown AE, Webster HK, Teja-Isavadharm P, et aL Macrophage activation in falciparum malaria as measured by neopterin and interferon gamma. Clin Exp Immunol 1990;82:97-101 26 Velde te AA, Hujbens JF, Heije K, et al. Interleukin4 inhibits secretion of IL-1i3, tumor necrosis factor, and IL-6 by human monocytes. Blood 1990;76:1392-7
Prevention of drowning Any map of Great Britain reveals a vast extent of coastline, river banks and lake shores. Moreover, as a seafaring nation there can be few people who at sometime or other have not been in, on, or on the edge of, water. So it is not surprising that there are a number of drowning fatalities. What is surprising is that, considering the total population, the numbers are so few. But it is extremely difficult, or impossible, to obtain accurate information about the circumstances surrounding drowning accidents beyond the bald statement that about 500-600 people die every year and that certain groups, eg the very young, the elderly and those taking part in water sports, carry a higher risk. In an attempt to obtain more information, the Medical Commission on Accident Prevention held a one day seminar at the Royal College of Surgeons of England to which interested parties were invited. They included: Royal Life Saving Society UK, Surf Life Saving Society,_ Central Council for Physical Recreation, Atlantic College, British Sub-Aqua Club, Royal Navy, Royal Air Force, St John Ambulance, Royal Society for the Prevention of Accidents, Consumer Safety Unit, Home Office, and the Department of Health. After much discussion it was agreed that in each risk group the events that lead to drowning were diverse and attempts at prevention needed to be approached in different ways. There was also agreement that the available data on the 'who' and 'why' were incomplete and in some cases very 'soft'. Accordingly, a small working party was set up under the Chairmanship of the Chairman of the Rescue and Resuscitation Committee of the Medical Commission on Accident Prevention consisting of a representative from the Royal Navy, the Royal Air Force, the Consumer Safety Unit of the Department of Trade and Industry, the Coroners Society, the Royal Life Saving Society and the British Sub-Aqua Club. At the outset it was found that there were no agreed figures for the annual number of deaths by drowning
453
27 Schindler R, Mancilla J, Endres S, et al Correlations and interactions in the production of interleukin-6 (IL-6), IL-1, and tumor necrosis factor (TNF) in human blood mononuclear cells: IL-6 suppresses IL-1 and TNF. Blood 1990;75:40-7 28 Troye-Blomberg M, Riley EM, Kabilan L, et al. Production by activated human T cells of interleukin 4 but not interferon -e is associated with elevated levels ofserum antibodies to activating malaria antigens. Proc Natl Acad Sci USA 1990;87:5484-8 29 Corelli F. Malarial therapy in Buerger's disease. Med World 1959;91:529-39 30 Yamada Y, Kimball K, Okusawa S, et al. Cytokines, acute phase proteins, and tissue injury. Ann NYAcad Sci 1990;587:351-61 31 Sinha AA, Lopez MT, McDevitt HO. Autoimmune diseases: the failure of self-tolerance. Science 1990;248:1380-7
in the UK. This was because a definition of found drowned does not always give a clue as to the possible cause. For example, many drowning deaths reported to the coroner may be suicide but because the victims are given the benefit of the doubt, their deaths are recorded as misadventure. Again, drowning accidents are sometimes reported more than once particularly at sea. One of the difflculties of collecting and collating facts about drowning is that there is no reliable body to check the details of an incident. The 'vital statistics' of drowning do not always give details of the circumstances and whether the victim was familiar with the environment. At present, the amount of detail about the deceased are dependent upon what is deemed necessary by the coroner. A postmortem examination is usually performed but detailed examination of, for example, the blood is not mandatory. In some states in Australia and North America tests for blood alcohol level are done on a routine basis and show a significant number of blood alcohol levels above the British legal limit. The Working Party was impressed by the inherent safety of organized groups professionally in contact with water, eg lifeboat crews, but was singularly unimpressed by the lack oftraining in those involved in recreational water activities. The excuse of the Department of Education is that the teaching curriculum is too full to include other- activities such as water safety. The Working Party could make no firm recommendations for improving water safety in adults without more information about drowning deaths. In view of the relatively small number of cases this could be said to be not cost-effective. But in children, particularly pre-school age, vigorous training would save a number of lives as well as teaching the invaluable discipline of 'think before you act'. B G-B Lucas Drowning Working Party The Medical Commission on Accident Prevention 35-43 Lincoln's Inn Fields London WC2A 3PN
0141-0768/91/ 080453-01/02.00/0 © 1991 The Royal Society of Medicine
Editorials 29 JUL 1991 Does malaria select for predisposition to autoimmune disease?
Over 20 years ago Dr Brian Greenwood observed that the admissions lists of a hospital in Nigeria contained very few cases of patients with autoimmune disease'. Of 98 454 admissions, only 104 were classified as autoimmune disease, including two with systemic lupus erythematosus (SLE) and 42 with rheumatoid arthritis, considerably less thanexpected (4- and 6-fold respectively) on the basis of figures for European hospitals. Greenwood suggested that parasitic disease, especially malaria, may in some way prevent the onset of autoimmune conditions. In contrast to its relatively low incidence in West Africans, SLE is more common in blacks (mainly descended from West Africans) than in whites in the USA2. The initial findings in West Africa were followed up by experimental work in which rodent malaria was shown to prevent the 'lupus-like' syndrome in one of the inbred strains of mice that spontaneously develop this symptom34 These interesting and straightforward observations have been largely ignored, probably because of the lack of a conceptual framework that could embrace the immune mechanisms involved in such complex events as malaria and autoimmunity. Recent advances in our understanding of the pathology of malaria and of factors that predispose to autoimmue disease (particularly SLE), suggest that a possible link between these conditions is tumour necrosis factor (TNF-aP. Although environmental factors are thought to precipitate SLE, there is a strong genetic prediosition in people with MHC class II DR2, DQwl and DR3 antigen alleles6. Mononuclear cells from subjects with DR2 and DQwl have a poor ability to produce TNF in vitro, suggesting-that this qytokine is involved in the pathogenesis of the disease6. However, SLE is also associated with DR3 aid mononuclear cells from donors with this specificity are good TNF producers6, the relationship is not a simple one. This is not surprising as SLE patients exhibit a broad spectrum of symptoms and the disease is considered to be an aggregate of similar clinical patterns rather than a single disease entity7. Nevertheless, it is significant that DR3 subjects do not exhibit the nephritis characteristic of other forms of SLE, while the DR2 and DQwl subjects do have this condition6. Further support for the argument that TNF- is involved in SLE-induced nephritis comes from experimental models: mice which are poor TNF producers and which are predisposed to lupus nephritis are by injections of TNF*. (Interestingly, a similar result is obtained with autoimmune diabetes9.) Malaria is a potent inducer of TNF in ai ,nals and man and it is now recognized that much of the pathology results from the initial induction of this cytokine'0. Patients infected with Plasmodium falciparum who develop very high plasma levels of so
TNF may be at greater risk from the potentially fatal condition of cerebral malariall. Malaria parasites express antigens (the so-called malaria 'toxins') which trigger macrophage release of TNF by T-cellindependent mechanisms'2, in a manner similar to endotoxin. Although the TNF-induced activation of macrophages and neutrophils enables the host to control parasite replication in the early stages,of the infection'3, there seems to be an uncontrolled overproduction of the cytokine that leads to a variety of pathological sequelae. This profound induction of TNF by malaria presumably explains why rodent malaria prevented the onset of autoimmune nephritis in susceptible strains of mice3'4 (and, possibly, adjuvant--nduced arthritis in rats'4). However, it would clearly be advantageous to the human host not to develop very high levels of TNF, thereby reducing the pathological consequences of malaria, and most especially the risk of cerebral symptoms. One might, therefore, expect to find that the long experience of malaria of populations in endemic areas, such as sub-Saharan Africa, would result in them exhibiting a lower level of symptoms than Caucasians. There is some evidence that this is indeed the case'5, although it has to be kept in mind that people from these areas may carry a variety, of inherited potective factors. It is conceivable that a consequence of a reduced ability to generate TNF in a malaria-free environment gives8rise to a greater risk of developing autoimmune conditions such as some forms of SILE - hence the higher rate of this in blacks in the USA (see above). It is well known that malaria may have acted as a selective agent for a variety of haemoglobinopathies etc., so it would not be suring if it has also selected genotypes involved in acute/non-specific immune responses. Unfortunately, our knowledge of the MHC genotypes of people in malaria endemic areas is rather limited, although the DQwl and DR2 haplotypes (see above) are slightly more commnn in blacks than in whites in the -USA (Hurley, personal-
communication). SLE, as already indicated, is a complex disease, considerably more common in women than men and usually with its onset during the main reproductive period. There may -be a further connection with malaria in this context, (although any attempt to explain the mechani involved at this stage has to be somewhat speculative). Even women who have become immune to malaria lose some (though not all) of this acquired resistnce during pregnancy and develop significant parasite burdens. The placenta becomes heavily parasitized, with inevitable consequences for the fetus in terms of reduced birth weight or possible abortion6L7. -However, the sytem by. stimulation of the mothefrs immu ye malaria may at the same time generate an increased output -of TNF and lessen the chances of SLE developing. Under natural conditions immunity to malaria is never complete, immune subjects always carry
0141-07/91/ 08041-W3=).00/0 © i99 The Royal Society of Medicine
452
Journal of the Royal Society of Medicine Volume 84 August 1991
parasites and suffer some fevers to a greater or lesser degree throughout their lives'8 so periodic minor boosts to the immune system are bound to occur and may be accompanied by rises in serum TNF19. It is especially pertinent to the argument that TNF administered to mice over a long period is immunosuppressive with respect to cell-mediated immunity (T cell proliferation, delayed hypersensitivity, etc.) but leaves humoral immunity largely unaffected20. The strain and species specificity of immunity to malaria, and the absence of delayed hypersensitivity to blood stage antigens21, suggests that naturally acquired resistance in humans is predominantly antibodymediated once it is established22, so the malaria itself is largely controlled and mortality is prevented. However, cell-mediated immunity, as exemplified by non-specific lymphocyte responses, is sometimes observed to be suppressed in subjects in endemic areas23. This mild general immunosuppression could perhaps contribute to preventing the loss of immune control characteristic of autoimmunity, by analogy with immunosuppressive drugs. Although malaria-induced TNF presumably contributes to the higher abortion rate of some mothers'7, (TNF induces abortion but a direct connection between malaria-induced abortion and TNF has yet to be confirmed) those generating low levels of TNF would have a greater chance of producing offspring, thus perpetuating their genotypes. Selection would be especially important at this point. Children in malaria-endemic countries carrying heavy burdens ofparasites without suffering apparent illness may develop antibodies to the parasite 'toxins' that directly stimulate TNF generation24, but it is also possible that their macrophages become tolerant to stimulation by these parasite components by analogy with endotoxin tolerance'3. Malaria could thus additionally select for the rapid development of control mechanisms that would shut down macrophage activation25 and TNF production, possibly through generation of other cytokines, particularly IL-426 and IL-627, as well as stimulating antimalarial antibody production28. In this context it may be significant that the genes for TNF map within the MHC complex6. Anecdotal evidence for malaria acting in an 'antiinflammatory' manner can be found in older literature. For example, P. vivax has been used to treat the gangrenous lesions of Buerger's disease29, thought by some physicians to result from an allergic sensitization to cigarette smoke; mention is also made of its use in the treatment of; other autoimmune problems29. Whatever the reliability of the data on Buerger's disease, a delicate balance may exist between the inflammatory and healing properties of cytokines, as proposed for IL-6 in the context of burns30, another condition in which acute nonspecific responses are generated. Although an understanding of autoimmune T cell reactivity3' is of major importance in explaining autoimmune disease (in addition to other factors, such as complement) a wider view that includes a focus onthe macrophage and TNF production, and important human parasites like malaria, may provide some useful clues that until now-have been ignored. Lastly, there are various grounds for thinking that a future malaria vaccine that allows the parasite to persist may be of more long-term benefit than one
which totally eliminates it from endemic areas. Members of these populations with a predisposition to autoimmune disease may be less disadvantaged by such a non-sterilizing vaccine. G A Butcher Department of Biology Imperial College of Science Technology and Medicine, Prince Consort Road, London SW7 2BB References 1 Greenwood BM. Autoimmune disease and parasitic infections in Nigerians. Lancet 1968;i:380-2 2 Lee SL, Siegel M. Systemic lupus erythematosus: epidemiological clues to pathogenesis. In: Dumonde D, ed. Infection and immunology in the rheumatic diseases. Oxford: Blackwell, 1976:307-17 3 Greenwood BM, Voller A. Suppression of autoimmune disease in New Zealand mice associated with infection with malaria. Clin Exp Immunol 1970;7:793-803 4 Greenwood BM, Herrick EM, Voller A. Suppression of autoimmune disease in NZB and (NZBxNZW) F1 hybrid mice by infection with malaria. Nature 1970;226:266-7 5 Butcher GA, Clark IA. SLE and malaria: another look at an old idea. Parasitol Today 1990;6:259-61 6 Jacob CO, Fronek ZL, Weis GD, et al. Proc Natl Acad Sci USA 1990;87:1233-7 7 Smolen JS, Chused TM, Leiserson WM et al. Heterogeneity of immunoregulatory T-cell subsets in systemic lupus erythematosus. Am J Med 1982;72:783-90 8 Jacob CO, McDevitt HO. Tumour necrosis factor-a in murine autoimmune 'lupus' nephritis. Nature 1988; 331:356-8 9 Satoh JO, Seino H, Shigeki S, et aL Inhibition of type 1 diabetes in BB rats with recombinant human tumor necrosis factor-a. J Immunol 1990;145:1395-9 10 Clark A, Chaudri G, Cowden WB. Role of tumour necrosis factor in the illness and pathology of malaria. Trans R Soc Trop Med Hyg 1989;83:436-46 11 Kwiatowski D, Hill AVS, Sambou I, et al. TNF concentration in fatal cerebral, non-fatal cerebral, and uncomplicated Plasmrnodium falciparum malaria. Lancet 1990;336:1201-4 12 Taverne J, Bate CAW, Sarkar DA, et al. Human and murine macrophages produce TNF in response to soluble antigens of Plasmodium falciparum. Parasie Immunol 1990;12:33-43 13 Clark IA. Cell-mediated immunity in protection and pathology of malaria. Parasitol Today 1987;3:300-5 14 Greenwood BM, Coller A, Herrick EM. Suppression of adjuvant arthritis by infection with a strain of the rodent malaria parasite Plasmodium berghei Ann Rheum Dis
1970;29:321-3 15 Powell RD, McNamara JV, Reickmann, KH. Clinical aspects of acquisition of immunity to falciparum malaria. Proc Helminth Soc Washington 1972;39(special
issue):51-66 16 Watkinson M, Rushton DI. Plasmodial pigmentation of placenta and outcome of pregnancy in West African mothers. BMJ 1983;287:251-4 17 McGregor IA. Epidemiology, malaria and pregnancy. Am J Trop Med Hyg 1984;33:517-25 18 Bruce-Chwatt LJ. A longitudenal survey of natural malaria infection in a group of West African adults. W Afr Med J 1963;12:141-73 19 Peyron F, Vuillez JP, Barbe G, et al Plasma levels of tumor necrosis factor during a longitudenal survey in an endemic area of malaria. Acta Trop 1990;47:47-51 20 Gordon C, Wofsy D. Effects of recombinant murine -tumor necrosis factor-a on immune function. JImmunol
1990*,l44:1753-8 21 Butcher~GA. In vitro responses of human peripheral blood mononuclear cells to Plasmodium fakciparum antigen. Int J Parasitol 1990;210:211-16
Journal of the Royal Society of Medicine Volume 84 August 1991 22 Butcher GA. Mechanisms of immunity to malaria and the possibilities of a blood stage vaccine: a critical appraisal. Parasitology 1989;98:315-27 23 Goonewardene R, Carter R, Gamage CP, et aL Human T cell proliferative responses to Plasmodium vivax: evidence of immunosuppression following prolonged exposure to endemic malaria. Eur J Immunol 1990;20:1201-16 24 Taverne J, Bate CAW, Playfair JHL. Malaria exoantigens induce TNF, are toxic and are blocked by Tindependent antibody. Immunol Lett 1990:25:207-12 25 Brown AE, Webster HK, Teja-Isavadharm P, et aL Macrophage activation in falciparum malaria as measured by neopterin and interferon gamma. Clin Exp Immunol 1990;82:97-101 26 Velde te AA, Hujbens JF, Heije K, et al. Interleukin4 inhibits secretion of IL-1i3, tumor necrosis factor, and IL-6 by human monocytes. Blood 1990;76:1392-7
Prevention of drowning Any map of Great Britain reveals a vast extent of coastline, river banks and lake shores. Moreover, as a seafaring nation there can be few people who at sometime or other have not been in, on, or on the edge of, water. So it is not surprising that there are a number of drowning fatalities. What is surprising is that, considering the total population, the numbers are so few. But it is extremely difficult, or impossible, to obtain accurate information about the circumstances surrounding drowning accidents beyond the bald statement that about 500-600 people die every year and that certain groups, eg the very young, the elderly and those taking part in water sports, carry a higher risk. In an attempt to obtain more information, the Medical Commission on Accident Prevention held a one day seminar at the Royal College of Surgeons of England to which interested parties were invited. They included: Royal Life Saving Society UK, Surf Life Saving Society,_ Central Council for Physical Recreation, Atlantic College, British Sub-Aqua Club, Royal Navy, Royal Air Force, St John Ambulance, Royal Society for the Prevention of Accidents, Consumer Safety Unit, Home Office, and the Department of Health. After much discussion it was agreed that in each risk group the events that lead to drowning were diverse and attempts at prevention needed to be approached in different ways. There was also agreement that the available data on the 'who' and 'why' were incomplete and in some cases very 'soft'. Accordingly, a small working party was set up under the Chairmanship of the Chairman of the Rescue and Resuscitation Committee of the Medical Commission on Accident Prevention consisting of a representative from the Royal Navy, the Royal Air Force, the Consumer Safety Unit of the Department of Trade and Industry, the Coroners Society, the Royal Life Saving Society and the British Sub-Aqua Club. At the outset it was found that there were no agreed figures for the annual number of deaths by drowning
453
27 Schindler R, Mancilla J, Endres S, et al Correlations and interactions in the production of interleukin-6 (IL-6), IL-1, and tumor necrosis factor (TNF) in human blood mononuclear cells: IL-6 suppresses IL-1 and TNF. Blood 1990;75:40-7 28 Troye-Blomberg M, Riley EM, Kabilan L, et al. Production by activated human T cells of interleukin 4 but not interferon -e is associated with elevated levels ofserum antibodies to activating malaria antigens. Proc Natl Acad Sci USA 1990;87:5484-8 29 Corelli F. Malarial therapy in Buerger's disease. Med World 1959;91:529-39 30 Yamada Y, Kimball K, Okusawa S, et al. Cytokines, acute phase proteins, and tissue injury. Ann NYAcad Sci 1990;587:351-61 31 Sinha AA, Lopez MT, McDevitt HO. Autoimmune diseases: the failure of self-tolerance. Science 1990;248:1380-7
in the UK. This was because a definition of found drowned does not always give a clue as to the possible cause. For example, many drowning deaths reported to the coroner may be suicide but because the victims are given the benefit of the doubt, their deaths are recorded as misadventure. Again, drowning accidents are sometimes reported more than once particularly at sea. One of the difflculties of collecting and collating facts about drowning is that there is no reliable body to check the details of an incident. The 'vital statistics' of drowning do not always give details of the circumstances and whether the victim was familiar with the environment. At present, the amount of detail about the deceased are dependent upon what is deemed necessary by the coroner. A postmortem examination is usually performed but detailed examination of, for example, the blood is not mandatory. In some states in Australia and North America tests for blood alcohol level are done on a routine basis and show a significant number of blood alcohol levels above the British legal limit. The Working Party was impressed by the inherent safety of organized groups professionally in contact with water, eg lifeboat crews, but was singularly unimpressed by the lack oftraining in those involved in recreational water activities. The excuse of the Department of Education is that the teaching curriculum is too full to include other- activities such as water safety. The Working Party could make no firm recommendations for improving water safety in adults without more information about drowning deaths. In view of the relatively small number of cases this could be said to be not cost-effective. But in children, particularly pre-school age, vigorous training would save a number of lives as well as teaching the invaluable discipline of 'think before you act'. B G-B Lucas Drowning Working Party The Medical Commission on Accident Prevention 35-43 Lincoln's Inn Fields London WC2A 3PN
0141-0768/91/ 080453-01/02.00/0 © 1991 The Royal Society of Medicine
